"Under Linux, many of the libraries are released as LGPL software, which allows non-Open Source software...to be compiled and linked to these programming libraries. This software then can remain as proprietary, non-Open Source software, even though it directly links to GPL software," the study pointed out, effectively killing the idea that the GPL has some kind of viral properties.
Er, the LGPL is not the GPL. It's a different license. The GPL does have viral properties, and that's the whole point of it.
Well I had a 486 firewall for a while with two ISA 10MBPS NICs. I got about 30KBps downloads with low (~30%) CPU usage. When I switched to a faster box on the firewall, I can now sometimes get around 200KBps. If it's not the ISA bus, I'd like to know what it is.
Is your firewall using an ISA network card? If so, that machine might be a bottleneck, even if the CPU isn't maxed out.
Re:I wonder if they really can make this 'invisibl
on
Foiling Cinema Pirates
·
· Score: 1
I just saw a show on the Discovery channel saying that digital cameras capture at 24 frames a second just to give the movie the same look as traditional film. To me that's like putting scratches and pops on a CD to make it sound like vinyl.
In fairness, when discussing Linux, bringing up the PDP-11 is not really fair, because Linux would be nowhere close to running on that machine. I don't think you can build a Linux kernel smaller than 300kb, let alone 64kb (though it might be fun to try).
But I take your point regarding how "unsuitable" the POSIX APIs are for small machines.
Higher fees won't stop the big companies, for whom (1) the actual filing fee is a small proportion of the total cost of a patent, and (2) patents are as good as cash in a world where the companies sign contracts with each other granting them permission to use patents en masse.
In contrast, the average citizen with a great idea and not a lot of cash lying around will be the one harmed by higher fees. In that case, I wonder, who are patents for?
Look at it this way: they are proposing to raise filing fees as a deterrent, and a "fee as a deterrent" is nothing but a fine. So, it seems now that ingenuity is an offence punishable by fine.
Only if you speak latin. While latin roots can be helpful, don't take them too literally. For instance, there's a portion of the brain called the "substantia nigra". By your logic, any black substance would qualify, but that is not the case.
Likewise, translucent and transparent have specific English meanings. The former means that diffusion of the transmitted light prevents the formation of a clear image. A thing is not transparent merely because it is "translucent enough". It's not a matter of how much light is transmitted---as the latin translation would imply---but rather a matter of how much diffusion occurs.
Man, I really need to stop posting about this now. Take a look at a dictionary, and if you still disagree, then that's your prerogative.
Translucent means that the object behind the material is distorted in some way.. such as darked or discolored.
That's not my understanding, and at least one dictionary disagrees with you:
translucent adj.
Transmitting light but causing sufficient diffusion to prevent perception of distinct images.
"Translucent" implies diffusion, not merely darkening or discolouration.
Coloured glass can be transparent or translucent depending on whether diffusion occurs.
I agree with Fred Brookes' comment that better languages can only help with accidental complexity, but can't remove the essential complexity of the problems we're trying to solve.
Where I disagree is on the relative proportions of these two things. I firmly believe that we're still at the point where well over half of what a programmer does fight with tools, rather than solve new problems. I'd go so far as to say I think it's about 90% accidental complexity, which implies that the ideal programming language could allow a given development team to produce systems that are ten times more elaborate/powerful/complex/etc than today's.
So I think the language designers still have a lot of work ahead of them.
I just think it might be a tad naive to think you could hit all the important test scenarios with a simple special-purpose functional test using malloc (though that is important for other reasons). You would need to be very insightful indeed to think of every possible thing that could go wrong in the Real World.
Instant messaging is becoming increasingly popular in the workplace. Local messaging servers like this were only the next logical step for businesses which don't wish to rely on an outside network for their messaging.
Yeah, it was the next logical step several years ago. IBM has been using the Sametime IM internally for as long as I have worked there.
Uh guys, they don't need hardware testing. I assume they trust that the RAM itself works. Besides, if all they wanted was "a few hours' worth of C", they could do that themselves.
They need to test the kernel. To do that, you need to run workloads that actually use more than 4GB of RAM, on a variety of hardware, or else you're not actually testing anything interesting in the kernel.
Slashdot Mirror
Well I had a 486 firewall for a while with two ISA 10MBPS NICs. I got about 30KBps downloads with low (~30%) CPU usage. When I switched to a faster box on the firewall, I can now sometimes get around 200KBps. If it's not the ISA bus, I'd like to know what it is.
See here.
Is your firewall using an ISA network card? If so, that machine might be a bottleneck, even if the CPU isn't maxed out.
I just saw a show on the Discovery channel saying that digital cameras capture at 24 frames a second just to give the movie the same look as traditional film. To me that's like putting scratches and pops on a CD to make it sound like vinyl.
But I take your point regarding how "unsuitable" the POSIX APIs are for small machines.
I hope a conscientious moderator will mod down my uninformed rants.
In contrast, the average citizen with a great idea and not a lot of cash lying around will be the one harmed by higher fees. In that case, I wonder, who are patents for?
Look at it this way: they are proposing to raise filing fees as a deterrent, and a "fee as a deterrent" is nothing but a fine. So, it seems now that ingenuity is an offence punishable by fine.
Likewise, translucent and transparent have specific English meanings. The former means that diffusion of the transmitted light prevents the formation of a clear image. A thing is not transparent merely because it is "translucent enough". It's not a matter of how much light is transmitted---as the latin translation would imply---but rather a matter of how much diffusion occurs.
Man, I really need to stop posting about this now. Take a look at a dictionary, and if you still disagree, then that's your prerogative.
Just last night I tried to watch a DVD that complained it was in the wrong region. Funny, the other three DVD in the set played without any problems.
See translucent and transparent.
Oh goody, I just can't get enough "in other news" posts. Bring them on.
The headline doesn't say they found a fault. To "find fault" in something just means to be critical of it.
Where I disagree is on the relative proportions of these two things. I firmly believe that we're still at the point where well over half of what a programmer does fight with tools, rather than solve new problems. I'd go so far as to say I think it's about 90% accidental complexity, which implies that the ideal programming language could allow a given development team to produce systems that are ten times more elaborate/powerful/complex/etc than today's.
So I think the language designers still have a lot of work ahead of them.
"inof"? Do you mean "in and of"?
Sorry... That's "Why would you send your resume in anything but PDF?"
Why would you send your resume in PDF anything but PDF?
We already use Lotus stuff internally.
I just think it might be a tad naive to think you could hit all the important test scenarios with a simple special-purpose functional test using malloc (though that is important for other reasons). You would need to be very insightful indeed to think of every possible thing that could go wrong in the Real World.
Cool. There's a language invented in 1970 called AIDS.
Give these guys some credit. They don't need volunteers to call malloc. They need real workloads.
They need to test the kernel. To do that, you need to run workloads that actually use more than 4GB of RAM, on a variety of hardware, or else you're not actually testing anything interesting in the kernel.
That's easy. I pick patents. That's what they're for.